Skip to content

Observed information matrix #30

@gkaissis

Description

@gkaissis

Is it possible to get the observed information matrix (i.e. the Fisher matrix, but just for every sample in a minibatch separately, not the expectation over the data distribution) by using your code? If yes, could you perhaps give a rough outline on how to accomplish this? Thanks!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions